This review article focuses on the extraction of drug-drug interaction (DDI) relations using deep learning techniques. DDI is crucial in drug development, pharmacovigilance, and patient safety. However, the small number of positive instances in existing datasets poses challenges for deep learning models to obtain sufficient feature information directly from text data. Therefore, the article introduces the general process of DDI relation extraction based on deep learning, summarizes various feature supplementation methods, and reviews state-of-the-art literature from a deep neural network perspective. The authors also compare the performance of different methods and discuss existing challenges and future research directions. The goal is to provide researchers with a comprehensive understanding of feature complementation methods to design and implement custom DDI relation extraction methods. The article is supported by grants from multiple institutions and is licensed under a Creative Commons Attribution International 4.0 License.This review article focuses on the extraction of drug-drug interaction (DDI) relations using deep learning techniques. DDI is crucial in drug development, pharmacovigilance, and patient safety. However, the small number of positive instances in existing datasets poses challenges for deep learning models to obtain sufficient feature information directly from text data. Therefore, the article introduces the general process of DDI relation extraction based on deep learning, summarizes various feature supplementation methods, and reviews state-of-the-art literature from a deep neural network perspective. The authors also compare the performance of different methods and discuss existing challenges and future research directions. The goal is to provide researchers with a comprehensive understanding of feature complementation methods to design and implement custom DDI relation extraction methods. The article is supported by grants from multiple institutions and is licensed under a Creative Commons Attribution International 4.0 License.